Question

我想使用机器学习算法测试日间交易逻辑。

我写了一个简单的代码，但没有工作，所以我需要你的帮助。

这就是：

逻辑

概念：在开盘当天买入S＆amp; P 500指数，在收盘日退出。
自变量：dayOpen/dayOpen(1)，dayOpen/dayClose(1)，dayOpen/dayHigh(1)，dayOpen/dayLow(1) ....... dayOpen/dayLow(2)

----＆GT; dayOpen(1) =前一天的开盘价
从属变量：if Close > Open分配1，否则为0
交易逻辑：使用具有8个功能的kNN或SVM算法，预测今天的索引方向。如果上涨，买入指数，否则持有现金。
我尝试用最近的60个滚动数据训练测试数据。

我的代码如下所示，但我收到了错误消息。有人可以帮帮我吗？

由于

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import pandas_datareader.data as web
from sklearn import neighbors, svm

def price(stock, start):
    price = web.DataReader(name=stock, data_source='yahoo', start=start)
    return price

clf1 = neighbors.KNeighborsClassifier(n_neighbors=3)
clf2 = svm.SVC()   

data = price('^KS11','2000-01-01')

data['Open(1)'] = data.Open/data.Open.shift(1)
data['High(1)'] = data.Open/data.High.shift(1)
data['Low(1)'] = data.Open/data.Low.shift(1)
data['Close(1)'] = data.Open/data.Close.shift(1)
data['Open(2)'] = data.Open/data.Open.shift(2)
data['High(2)'] = data.Open/data.High.shift(2)
data['Low(1)'] = data.Open/data.Low.shift(2)
data['Close(2)'] = data.Open/data.Close.shift(2)
data['Result'] = np.where(data.Close/data.Open>1,1,0)
data['Close/Open'] = data.Close/data.Open
data['Prediction'] = pd.Series()

for i in range(61, len(data.index)):
    x = data.iloc[i-61:i,6:14]
    y = data.Result[i-61:i]
    clf2.fit(x,y)
    data.Prediction[i] = clf2.predict(x)[-1]

data = data.dropna()
data['Profit'] = np.where(data.Result == 1,data['Close/Open'],1).cumprod()
print(data.iloc[:,6:])
data.Profit.plot()
plt.show()

ValueError                                Traceback (most recent call last)
<ipython-input-8-8c10864c79e2> in <module>()
     29     x = data.iloc[i-61:i,6:14]
     30     y = data.Result[i-61:i]
---> 31     clf2.fit(x,y)
     32     data.Prediction[i] = clf2.predict(x)[-1]
     33 

C:\Users\yiugn_\Anaconda3\lib\site-packages\sklearn\svm\base.py in fit(self, X, y, sample_weight)
    148         self._sparse = sparse and not callable(self.kernel)
    149 
--> 150         X = check_array(X, accept_sparse='csr', dtype=np.float64, order='C')
    151         y = self._validate_targets(y)
    152 

C:\Users\yiugn_\Anaconda3\lib\site-packages\sklearn\utils\validation.py
in check_array(array, accept_sparse, dtype, order, copy,
force_all_finite, ensure_2d, allow_nd, ensure_min_samples,
ensure_min_features, warn_on_dtype, estimator)`enter code here`
    396                              % (array.ndim, estimator_name))
    397         if force_all_finite:
--> 398             _assert_all_finite(array)
    399 
    400     shape_repr = _shape_repr(array.shape)

C:\Users\yiugn_\Anaconda3\lib\site-packages\sklearn\utils\validation.py
in _assert_all_finite(X)
     52             and not np.isfinite(X).all()):
     53         raise ValueError("Input contains NaN, infinity"
---> 54                          " or a value too large for %r." % X.dtype)
     55 
     56 

ValueError: Input contains NaN, infinity or a value too large for dtype('float64').

需要帮助python

0 个答案: